Payday Super receipt evidence

Receipt dates and amounts

Checker 0.1.7 requires amount evidence for a timely receipt. Verified on 22 September 2026: a $120 obligation with a timely receipt date but neither matched_amount nor remitted_amount returns UNKNOWN and exit 2. A late receipt without an amount remains LATE, but cannot reduce the shortfall and uses the as-at date for the maximum estimate. Read the 0.1.7 release record.

Older checkers 0.1.5 and 0.1.6 assume full receipt when that amount is missing. The 0.1.6 release discloses the assumption in a caveat. Published MCP 0.2.8 pins checker 0.1.7 and includes the correction; MCP 0.2.4 pins 0.1.6 and retains the older behaviour.

matched_amount is the supplied amount associated with the payday. It takes precedence over remitted_amount, the supplied amount sent. Neither field authenticates a fund record. A remitted amount without fund receipt evidence does not establish receipt.

For the $120 obligation in this example, a timely $60 matched receipt returns UNPAID, not full payment. A timely $120 matched receipt returns ON_TIME only on the supplied facts. Late receipt returns LATE; a receipt date before the remittance date is rejected. Resolve unmatched records and contradictory evidence before relying on any result.

Workbook versions and limits

The tagged 0.1.6 workbook assumes full receipt when a receipt date has no amount. It also extrapolates GIC beyond its last covered day, 31 December 2026. The 0.1.6 CLI withholds those estimates by default and offers --allow-stale-gic as an explicit opt-in; the workbook has no equivalent option.

The tagged 0.1.7 workbook fixes both cases. Its Summary shows Not assessed for rate-dependent totals if any row needs an uncovered rate. It is available from the 0.1.7 tag; a PyPI install does not install the workbook. Desktop Excel checks on 22 September 2026 (Australia/Sydney) covered receipt amounts, remittance acknowledgements, contradictory dates and both sides of the GIC boundary. The verification record identifies the tested files and hashes. These checks do not establish parity for every input or option. Read Review Checks before using row results; invalid input can leave a displayed verdict while the workbook is BLOCKED.

Try the CSV without cloning

With uv installed, download the synthetic CSV as timely_remittance_no_receipt.csv. Open a terminal in that folder and run:

uvx --from payday-super-checker==0.1.9 payday-super-check timely_remittance_no_receipt.csv --as-at 2026-08-20 --confirm-remittance-only -o payday-report.csv

Quick trial release 0.1.9, the current published release on 1 October 2026. The command downloads checker 0.1.9 from PyPI and writes payday-report.csv, replacing that file if it exists. Open it to find one AT_RISK contribution with due date 2026-08-17.

Check the matching evidence before relying on the exported file. If employee identifiers or pay period columns are missing, the join degrades and which super payment matches which payday can change. Checker 0.1.5 carries each structural matching warning into the report’s caveats column; checker 0.1.4 printed them to the console only, so keep that console output with an older report. Reconcile affected matches before using a verdict; a row labelled ON_TIME is not reliable evidence of the match itself. See Known limitations (PSC-2).

Exit 0 records your remittance-only acknowledgement. It does not establish timely receipt. Omit --confirm-remittance-only to get the same verdict with exit 2 and a request to resolve or acknowledge the evidence gap.

The fixed example above and the recorded public evaluation use release 0.1.3. Follow that evaluation’s revision and commands to reproduce the historical run. This quick trial uses release 0.1.9 and leaves the fund receipt decision with the reviewer.

Evidence pack workflow

The CLI change combines the existing checker and practitioner checklist in one command. It writes report.csv, practitioner-review.md, exceptions.json and a blank decision-log.md into a new directory.

From checker 0.1.5 the pack’s report carries the structural matching warnings described under Known limitations (PSC-2). Checker 0.1.4 printed them to the console only, so keep its transcript with an older pack. Reconcile affected matches before relying on any verdict.

The report omits employee identifiers and input paths. The checklist and exception queue bind to its SHA-256. Every result other than ON_TIME enters the queue; the fictional example above remains AT_RISK and exits 2. A remittance-only acknowledgement does not clear that queue.

A reviewer uses source row references to reconcile the private input, obtains missing receipt evidence and records decisions under applicable APES 110 and TPB obligations. Existing decision logs are preserved by refusing an output directory that already exists. The files remain private workpapers containing dates, amounts and warnings.

The companion MCP tool, build_payday_super_evidence_pack, accepts structured contribution rows and returns the 4 files in memory. It accepts no input or output paths, writes no files and records no decisions. It requires a checker with evidence pack support.

The identifier-free report is for the included checklist. Existing review-pack and PaydaySuper.Report Excel imports require the ordinary checker’s 18-column report.

This workflow is included in published checker 0.1.4 and MCP 0.2.2. Checker 0.1.3, used by the historical evaluation, does not include it. Reproduce the recorded development snapshot with fictional data. No advice, lodgement or ledger write is performed.

Check the evidence

Reproduced from Payday Super Checker 0.1.3. The release’s legal review date is 15 August 2026.

The ordinary timing pathway tests fund receipt within 7 business days of payday. Check whether a different timing pathway applies before relying on the result. Primary sources are the Superannuation Guarantee (Administration) Act 1992 and Australian Taxation Office, LCR 2026/2, as recorded in the evaluation pack.

Source on GitHub · Open an issue

Alongside the receipt amount limitation, two further limitations are recorded against this tool: an extrapolated general interest charge rate past the end of the published table, and the ways the payroll-to-super join degrades when identifiers do not match. See Known limitations.